Official statistics show more deaths in patients admitted to hospital at weekends

BMJ 2016; 355 doi: (Published 27 October 2016) Cite this as: BMJ 2016;355:i5797
  1. Nigel Hawkes
  1. London

Patients admitted to an NHS hospital in England at the weekend in 2015-16 were 15% more likely to die within 30 days than those admitted during the week, an official analysis of the statistics shows.1

Furthermore, those discharged at the weekend were more likely to be readmitted as an emergency within seven days, and Sunday discharge was associated with a 40% higher readmission rate than discharge on a Wednesday. Friday and Saturday discharges were also linked to increased readmission rates, by 9% and 27%, respectively.

The mortality rates calculated by NHS Digital (formerly the Health and Social Care Information Centre) were very similar to those of Nick Freemantle and colleagues published in The BMJ,2 which is not surprising, as the same methodology was used. The readmission data were new, and NHS Digital has produced tables listing the outcomes of both analyses among all trusts, excluding those with fewer than 100 deaths or readmissions. The data originated from the Hospital Episode Statistics database.

NHS Digital also analysed the length of stay by the day of admission, but this did not show any striking changes. Although the median length of stay among people admitted on a Friday, Saturday, or Sunday was two days, as opposed to one day among all other admission days, NHS Digital warned against drawing conclusions from this.

A trust by trust comparison of mortality and readmission shows considerable variation. Overall, 58% of trusts show a statistically significant increase in mortality among weekend admissions, and 12% show a significant increase among admissions on Mondays and Fridays, the transition period between the week and weekend. By including only emergency admissions rather than all admissions, the risk of dying increases by 12% rather than 15%.

Among emergency readmissions, 85% of trusts show a significantly raised odds ratio for Sunday admissions and 79% for Saturday admissions. Friday admissions were linked to a higher risk of readmission in 37% of trusts.

Delving into the tables shows that the trusts with higher odds ratios of mortality after weekend admissions in 2015-16 included Morecambe Bay (odds ratio 1.35 (95% confidence interval 1.20 to 1.53)), South Warwickshire (1.41 (1.21 to 1.65)), and South Manchester (1.35 (1.20 to 1.53)). The Royal Brompton and Harefield trust had one of the highest (1.48 (1.15 to 1.9)).

NHS Digital also included data from 2013-14 and 2014-15, which showed very similar overall results but big variations between individual trusts. Royal Brompton showed no significant increase in 2013-14, for example (1.22 (0.95 to 1.58)), and does not figure in the 2014-15 data because deaths fell below 100. Morecambe Bay similarly showed no significant increase in 2013-14 and a much smaller one in 2014-15 (1.21 (1.06 to 1.39)).

NHS Digital warns against drawing too much from these statistics because of the many possible explanations, including differences in case mix. The episode statistics include no measure of severity of the cause of admission, and more acute cases may tend to be admitted at weekends.

Hence, the publication of these statistics is unlikely to end controversy over whether a seven day NHS would reduce mortality and readmissions. Those who believe that it would—including the health secretary, Jeremy Hunt—will find nothing to disprove their assertions.

Mark Porter, BMA chair of council, said, “Collecting and analysing figures such as these is always important in improving patient care. We should remember, however, that these figures are not a research study and can provide only a description of differences, not the reason for any differences.”

He added, “We should not forget that, over the past year, a long line of healthcare professionals and leading experts have challenged the misleading use of mortality figures for political ends. The underlying picture is always much more complicated than bare figures show.

“The BMA believes that patients should have access to the same high quality of care, seven days a week, and doctors already work around the clock delivering care for patients. Crucially, if the government wants to make more services available across seven days, then it needs to explain how it will staff and pay for them at a time when existing services are struggling to keep up with demand.”


View Abstract

Log in

Log in through your institution


* For online subscription